Strategic Management for Credit Risk in Supply Chain Networks: A Novel Framework

Abstract

This study addresses the imperative for robust credit risk management strategies by proposing a novel framework tailored for supply chain networks. It aims to bridge existing gaps in credit risk assessment methodologies by amalgamating empirical insights, advanced computational techniques, and comprehensive data analytics. Leveraging a comprehensive dataset encompassing diverse attributes crucial for credit risk assessment, this study employs a meticulous methodology. It integrates machine learning algorithms, notably LightGBM, and exploratory data analysis techniques to preprocess data, examine missing values, assess variable correlations, and construct a predictive model. The empirical journey reveals insightful findings, emphasizing missing value patterns, variable interrelationships, and model performance. Precision-recall and ROC curves depict the model's ability to discern default and non-default cases, showcasing its efficacy in credit risk assessment within supply chain contexts. Our study contributes a foundational framework for strategic credit risk management within supply chain networks, offering actionable insights for stakeholders. While acknowledging limitations and the need for ongoing model refinement, this research sets the stage for future explorations and transformative practices in adaptive risk management strategies for interconnected supply chain networks.